a post-kaleckian, post-olsonian approach to unemployment and...

ANNALS OF ECONOMICS AND FINANCE 2, 357–383 (2006)

A Post-Kaleckian, Post-Olsonian Approach to Unemployment

and Income Inequality in Modern Varieties of Capitalism

Arne Heise*

Department of Politics and Economics, Hamburg University, VMP 9E-mail: [email protected]

The ‘big trade-off’, described by Arthur Okun some thirty years ago, is backagain. Equality or efficiency, or to put it differently again: modern highly de-veloped economies and societies must choose between the Scylla of incomeinequality or the Charybdis of unemployment. Furthermore, it seems thatthe continental European economies — foremost Germany and France — havesided with more egalitarian ends accepting higher unemployment. This has oc-curred whilst the liberal economies such as the United States and the UnitedKingdom choose higher inequality for lower unemployment. In this paper it isargued that the trade-off is not a supply-side necessity to maintain work effortin a situation of incomplete contracts, but is a politico-economic issue betweenparticular interest groups to seek rents. However, unlike in Mancur Olson’sseminal approach, it is not the trade unions which are forming distributionalcoalitions on the labour market. Rather, it is the meritocracy who are happyto use Keynesian-type demand management in order to advance their mate-rial interests by pursuing a ‘Meritocratically Optimal Rate of Unemployment’(MORU). c© 2006 Peking University Press

Key Words: Unemployment; Income inequality; Political economy.JEL Classification Numbers: E12, E24, E25, P16, P51.

1. INTRODUCTION: THE AGE OF INEQUALITY

The quest for equality — as a means to bolster against poverty and asthe expression of social justice — has been strongly attacked as ‘egalitar-

* This paper was written while the author was a Visiting Fellow at St. Edmund’sCollege of the University of Cambridge in June-July 2005. It is part of a larger researchproject ‘Employment systems in international comparison’ financially supported by theHans Bockler Foundation to whom the author is very grateful. He is also grateful toPhilip Arestis for comments and to Michael Murphy for corrections.

3571529-7373/2006

Copyright c© 2006 by Peking University PressAll rights of reproduction in any form reserved.

358 ARNE HEISE

ianism’ that destroys market incentives1 and as being incompatible withfull employment in the age of the service economy. Indeed, the issue wasaddressed as early as the 1970s by Arthur M. Okun2 under historical cir-cumstances completely different from today’s: In the 1970s US incomedispersion was decreasing, while unemployment was higher than in mostEuropean countries. Furthermore, the world was divided into two different,hostile ideological camps of capitalism and communism. At least from anideological point of view in the battle of systems during the Cold War, thealleged higher efficiency of the capitalist market economy was not to bepaid for by lower equality as this would leave the question of superiorityopen to different preferences. Captialism could be proven as superior overcommunism only if it could combine a sufficiently high degree of equality(social justice) with superior economic efficiency (in terms of growth andemployment). This would highlight the notorious failure of communismto support (intensive) growth. Moreover, the economies of the Europeancountries — being extremely close to the ideological divide — seemed to beable to fulfil these promises during the 1960s and 1970s particularly well.

Only after the downfall of the Berlin Wall in 1989 and the end of the sys-tems confrontation, the need to combine equality and efficiency has finallydisappeared. Interestingly, the renewed (alleged) trade-off between equalityand efficiency (in terms of providing high levels of employment) unites theliberal-conservative political and academic camp with most of the parts ofthe progressive, social democratic political and academic camps. However,whilst the former welcomes the implicit demand for higher inequality3, thelatter somewhat hastily demands for a (albeit hard) choice.4 It is in thissense that we can speak of the age of inequality: an undisputable trendwith undisputable negative consequences (the correlation between raisinginequality and growing poverty). It is an ideological issue (‘social justice’)that centres prominently in political discussions and electoral campaigns,and is an academic discourse that focuses completely on the microeconomiclogic of market incentives. Very few dissenting economists have tried tokeep a macroeconomic eye on the issue5 and — to my knowledge — therehas not yet been a consistent political economy approach focusing on theobvious material interests involved.

1This incentive problem has been put as: ‘The poor reduce their labour supply becausethey earn too much and the rich withdraw their labour supply because they earn toolittle’. John Kenneth Galbraith’s reaction was said to be: ‘Who believes this, will believeanything’.

2See Okun (1970), Okun (1975).3See e.g. Machan (2002), Davis/Meyer (2000), Henry (1995), Kekes (2003), Gier-

sch/Paqu/Schmieding (1992) and, most infamously, Gilder (1982).4See e.g. Iversen (1999), Scharpf (2001) and Schettkat/Appelbaum (1996).5Most prominently Galbraith (1998) and Harrison/Bluestone (1990).

A POST-KALECKIAN, POST-OLSONIAN APPROACH 359

This is where this paper comes in: I will try to explain a different un-derstanding about the evolution of unemployment and income inequalityover the past two decades. I will do this by taking a political economyapproach, i.e. which will centre upon the different distributional interestsamong societal groups in different capitalist economic settings. In parttwo, I will extend Mancur Olson’s seminal approach to the political econ-omy of economic development by focusing on the labour market and bygiving Olson’s approach a Keynesian twist. In part three, I will refer toMichal Kalecki’s famous proposition that unemployment can be used asa strong mechanism to shift power relations in the distributional conflictof capitalism. However I will transform Kaleckis’s proposition into a de-terminant of personal rather than functional income distribution. In partfour, I will then join the different proposition strings together and derivean unemployment situation which may be called optimal as it best servesthe interests of those largely capitalising from pure market outcome — themeritocrats. Furthermore, it will become apparent that such a ‘Meritocrat-ically Optimal Rate of Unemployment’ (MORU) may be very diverse overthe range of varieties of capitalistic models. In part five, the predictionsof the MORU theory will be empirically tested. As we are dealing withvested interests often transmitted into political action through ideologicalinfiltration, the carrier of the message himself is sometimes accused of anideological bias6 — that is, why any empirical test of political economyissues is so important.

2. MANCUR OLSON, DISTRIBUTIONAL COALITIONSAND ITS KEYNESIAN TWIST

To apply a Keynesian approach to political economy seems both neces-sary and fruitful in order to provide the ‘missing link’ in Keynesian eco-nomics: “Keynesian economists had failed to develop a general explanationof governments’ unwillingness to supply the aggregate demand policies re-quired to achieve full employment”, as John and Wendy Cornwall (2005:6) claim.

Mancur Olson provides the starting point as he has established one ofthe most powerful political economy approaches in his ‘Rise and Declineof Nations’7 that directly links income inequality and unemployment. Heclaims that societies produce ever more interest groups that seek to advancetheir particular well-being at the expense of the well-being of the societyas a whole. Such interest groups are only able to pursue this kind of ‘rent-seeking’ behaviour — which comes as a public good to all individuals who

6Obviously, ‘ideology’ as an object of scientific investigation is here confused with‘ideology’ as a subject guiding (un)scientific investigation.

7Olson (1982)

360 ARNE HEISE

comprise as a group — if the groups are small and homogenous enoughto overcome the ‘cooperation trap’.8 Although such interest groups mayevolve in every market9, according to Olson, unemployment is best ex-plained if interest groups on both sides of the labour market are takeninto focus. These are trade unions and employer’s organisations. For tworeasons, in a variety of microeconomic approaches (insider-outsider theo-ries, NAIRU theories, efficiency bargain and right-to-manage models) thesupply side has been set at centre stage of the investigations: Firstly, un-employment is usually associated with the notion of ‘too high wages’ whichcan better be explained if labour supply (by forming unions) is able tocollude an effective distributive coalition. Secondly, once we assume com-modity markets as competitive, there would always be pressure to raisereal wages by reducing prices — which ought to happen if Walras’ Law ofmarkets applies. Therefore, if unemployment becomes a long-term problemof advanced economies, according to Olson, it must stem from the willing-ness and ability of trade unions to effectively seek rent for their members.These members are mostly low skilled, blue collar workers in manufacturingindustries. Hence, the variance in national labour market performances ob-viously depends on the incubation process of interest groups in general andtrade unions in particular throughout the time period without significantinstitutional, legal or other alterations in a particular society10. The longerthis time period, the more distributional coalitions, the lower growth andhigher unemployment will be. Olson’s approach can therefore be describedas a theory of ‘societal sclerosis’.

Olson’s central idea that the pursuit of vested distributional interestsmay impinge on the general welfare of society, is a challenging one. How-ever his focus on the supply side of the labour market is too narrow. Onthe one hand, he underestimates what has been called ‘the radius of trust’in the literature on social capital, i.e. the possibility and willingness ofgroups (trade unions in this case) to internalise the external effects of theiractions in order to reduce the negative repercussions of rent-seeking be-haviour.11 Secondly, the reasoning rests entirely on the microeconomicsof general equilibrium theory. However, if Keynesian macroeconomics are

8This has been established and elaborated in his famous ‘Logic of Collective Action’;see Olson (1965).

9“Such action occurs through professional associations, labour unions, farm organiza-tions, trade associations, and oligopolistic collusions of firms in concentrated industries”(Olson 1996: 74).

10Olson (1996: 76) mentions ‘revolutionary upheavals’, ‘totalitarian repression’ or‘foreign occupations’. However, drastic institutional change (e.g. under the Thatcherera) may also be counted.

11Of course, Olson mentions ‘encompassing organisations’, which seem to fit the re-quirements of a wide radius of trust. However, Olson firmly believes that these encom-passing organisations are unstable and necessarily dissolves eventually.


to be discussed, unemployment (at its core) may no longer be the resultof misguided wage policy but of macroeconomic policy failures. Further-more, here, income inequality (in an equilibrium position) would no longerbe technically determined by the marginal productivity of different jobsrequiring different skills. Alternatively it would depend on what societyrates as ‘fair’ and ‘tolerable’. This is something James K. Galbraith (1997;1998) calls ‘job or wage structure’.12 Here, of course, unemployment canbe used in order to change established views about ‘fair’ and ‘tolerable’levels of income dispersion, if Keynesian-type macroeconomic policy po-tentials are not sufficiently used to cure unemployment when it occurs or,even worse, are redirected to create or maintain unemployment. Now, inthis post-Olsonian perspective, it would be the demand-side of the labourmarket, i.e. employers and their organisations, that pursues its vested dis-tributional interests by wasting capacities and reducing general welfare forthe society as a whole.

3. MICHAL KALECKI, UNEMPLOYMENT AND INCOMEDISTRIBUTION, OR: THE MERITOCRATICALLYOPTIMAL RATE OF UNEMPLOYMENT (MORU)

As John and Wendy Cornwall (2005) claims, most Keynesians13 failed inposing the question of political limits14 to the practice of Keynesian-typeinterventionist employment policies. Yet, this may be at least partly dueto the fact, that Keynes himself was not very thoughtful (if not downrightnaıve) about the possible impact of vested interests on economic policy.Joan Robinson (1976) noted:

“Keynes liked to believe in the power of ideas to influence the course ofhistory. He sometimes maintained that when the principles of employmentpolicy were understood, economic affairs would be conducted rationally, and heeven went so far as to predict a happy future in which our grandchildren coulddevote themselves entirely to the arts and graces of life.”

Robinson also singled out whom she believed to be closer to the realworld:

“Kalecki’s vision of the future was more realistic. In a remarkable articlepublished in 1943 on the ‘Political Aspects of Full Employment’ he foresaw

12“It is a historically, socially, and politically specific set of values and pay relation-ships in the economy, within and between firms, within and across industries” (Galbraith1997: 15); see also Galbraith (1998:50ff.).

13Particularly Philip Arestis and Malcolm Sawyer (1998) must be explicitly excludedfrom this charge.

14In a mini-symposium on ‘Is Keynesian demand management policy still viable?’published in Vol. 17 (1994-95) of the Journal of Post Keynesian Economics, not evenone of the papers mentions political limits. There may also be economic limitations dueto the lack of cooperation of different policy actors; see Heise (2001), Heise (2005a).

362 ARNE HEISE

that when governments understood how to control the commercial trade cyclewe should find ourselves in a political trade cycle.”

Actually, Michal Kalecki (1943/1990) had not — as is often claimed —described a ‘political business cycle’15 but pointed out that ‘the industrialleaders’ may not be interested in full employment altogether. As Kaleckiwas not engaged in the investigation of the connections between the econ-omy, economic policy and the political and electoral system, there is no hintof a political business cycle theory16 but rather of a ’political equilibriumtheory of unemployment’ and an indication of how this will be achieved.This is through mystifying an active (i.e. deficit-financed) budgetary pol-icy as ‘unsound’ and ‘perilous’ (see Kalecki 1943/1990: 350). A situationof lasting full employment would be, from the perspective of the ‘captainsof industry’17, unwarranted on three accounts: 1) In general, they do notwant the government to interfere with market processes; 2) they do notwant the government to spend money in areas — such as public invest-ment — where there is a potential competition with or crowding out ofprivate investment; 3) they feel that lasting full employment changes thepower relations towards the working classes and takes away an effective dis-ciplinary device. These rather cloudy expositions have been re-framed inmodern, orthodox labour market theory: In efficiency wage theories, unem-ployment guarantees to maintain productivity growth. In NAIRU theories,unemployment is needed to prevent the acceleration of inflation. Finally inheterodox, Kaleckian macroeconomics, full employment destroys the socialstructure of accumulation by squeezing profits18. In equilibrium, in orderto appropriate profits, capitalists need unemployment or workers will beable to claim any surplus for themselves (in any combination of goods andleisure).19

15Politico-economic models and cycles have been introduced in the literature byWilliam Nordhaus (1975) and R. Boddy and J. Crotty (1974) who refer to MichalKalecki’s work.

16Although he claims to have presented ‘a regime of the political business cycle’; seeKalecki (1943/1990: 355). However, he uses the term ‘political business cycle’ in acompletely different way than it has become customary since Nordhaus; see e.g. Henley(1988).

17Kalecki uses different descriptions for those, who pursue their special intereststhrough lasting unemployment: ‘captains of industry’, ‘industrial leaders’, ‘bosses’,‘business leaders’, ‘businessmen’, ‘big business’. This leaves us without a clear idea ofwhom he really envisages: capitalists as the class of capital owners and representativesof profit-earners in functional income distribution or executive and high-ranked whitecollar employees (and also profit-earners) as representatives of the higher income per-centiles in personal income distribution. From this context and later work, one can inferthat Kalecki is referring to capital owner and, therefore, functional income distribution.

18See e.g. Kotz/Donough/Reich 1994; Marglin/Schor 1990;Bowles/Edwards/Roosevelt 2005.

19Henley (1988: 439) points to the parallels to Marx’s ‘reserve army’.


Without engaging in fundamental discussions about functional incomedistribution, this Kaleckian version of a political economy of unemploymentcannot be made easily compatible with two decades of post-war economichistory. The co-existence of full employment and a positive profit sharein most OECD countries in the 1950s and 1960s would either require theassumption of a disequilibrium situation for the entire period or some de-gree of monopolistic competition in almost very commodity market. Or,as in Post Keynesian macroeconomics, functional income distribution (andthe existence of profits) does not depend on unemployment but rather ona positive rate of interest (which closes the only degree of freedom of thedistributional system; see e.g. Sraffa 1960) which is determined by liquiditypreference considerations (see e.g. Riese 1981). However, these qualifica-tions do not necessarily eradicate the Kaleckian idea altogether, but mightshift the focus of intention from functional to personal income distribution,and from class conflict to meritocratic deliberations. As the expression‘meritocracy’ is far from being clearly defined, let us refer to the FontanaDictionary of Modern Thought (1977: 384):

“A word coined by Michael Young (The Rise of Meritocracy, 1958) for govern-ment by those regarded as possessing merit; merit is equated with intelligence-plus-effort,... Egalitarians often apply the word to any elitist system of educa-tion and government,...”

Moreover, according to Benabou (2000: 321), there is “no single, value-free definition of meritocracy, but only preference orderings about equalityof opportunities and equality of outcomes”. Meritocracy, therefore, mustbe placed between the one extreme of aristocracy, where opportunities andoutcomes are structured only by decent, and the other extreme of egalitar-ianism, where opportunities and outcomes are entirely equally distributedamong the members of a society. Obviously, modern societies have aban-doned aristocratic structures and float somewhere between meritocracy andegalitarianism. Despite the degree varying considerably through redistribu-tive welfare arrangements and institutions directly impinging on marketoutcomes, in every highly developed country some measure of egalitarianredistribution of opportunities and outcomes are employed.20 Thereforethe concept of meritocratic deliberations used for our purpose puts thequest for income inequality at centre stage: in line with economic merits— defined by ‘pure’ market outcomes accepting an initial endowment ofreal, financial and human capital — any increase in net income dispersionis welcomed as improving economic efficiency. Obviously, someone who isbetter endowed with real, financial or human capital would profit materiallyfrom any meritocratic policy shift — yet, I do not purport that the better-

20Although welfare arrangements have not been historically invented for egalitarianpurposes, they do have this effect; see e.g. Baldwin (1990).

364 ARNE HEISE

offs — or as I will call them in line with the above given description: theelite21 — necessarily pursue meritocratic aims, as I am not claiming thatthe non-elite necessarily favours egalitarian ends. This may, for instance,vary considerably with the individual preference structure (materialistic,post-materialistic) or individual perception of income mobility. However,elite studies reveal a strong meritocratic orientation of the elite which hasbecome more pronounced over the past two decades.22

To summarise, the above outlined post-Kaleckian version of the politicaleconomy of unemployment does not focus on the class struggle over func-tional income shares but puts forward the idea that unemployment may be(mis-)used by the elite in order to champion meritocratic ideas of personalincome distribution. For reasons having been advanced by Lester Thurow(1971) and which we will discuss below, the elite may not want to end re-distribution and a compression of the income (and particularly the wage)structure entirely, but seek an ‘optimal degree of income inequality’. Orto put it differently: the elite are tempted to address a policy pursuing a‘Meritocratically Optimal Rate of Unemployment’, abbreviated as MORU.

4. MORU IN THE VARIETIES OF CAPITALISM

Many studies have shown that unemployment is a strong and significantfactor in determining income inequality in general and wage dispersion inparticular as measured by a gini-coefficient or by the ratio of highest tolowest income decile (P90/P10).23 Here, nominal wage increases in rela-tion to labour productivity growth and (expected) inflation — i.e. the‘distributional margin’ — will come under pressure if wages are to play

21‘Elite’ is another ill-defined notion. Primarily, it is used for those groups of individ-uals that are able to take or influence collective rather than merely individual decisions:political, economic, organisational or media elites for instance. I presuppose that thereis a strong correlation between being materially better-off and being part of the elite -and that is why I use ‘elite’, ‘better-offs’ and ‘meritocrats’ interchangeably.

22This is definitely true for the media, economic and liberal-conservative politicalelites in Germany. It is also true for the social-democratic political elite that havebecome markedly more meritocratic over the past two decades. Therefore, it is basicallythe trade union elite that still takes a non-meritocratic, egalitarian perspective; see e.g.Brklin (1997).

23See e.g. Forster (2000) or Volscho (2004) for a comprehensive overview.Jantti/Jenkins (2001) take a more sceptical view — yet they do not deny the rela-tionship but point to a more complex interaction between unemployment, inflation andinequality. Below, I will discuss in more detail which income variable to be taken inorder to capture the inequality visions of the meritocracy. Suffice to say here, that wagedispersion will serve as proxy as wage income is still by far the most important source ofincome for the vast majority of the people and the link between wage income and unem-ployment (both being determined on the labour market) is the closest. Furthermore, ithas been shown that recent trend in income inequality largely stem from the underlyingtrends in wage dispersion; see Forster (2000: 8).


any role in signalling relative scarcity. This is one of the best establishedempirical relations in economics that lies behind the (original) ‘Phillipscurve’. Here, on the one hand it will, at least temporarily (as long as dis-equilibrium lasts), shift functional income distribution in favour of capitalowners — who may be supposed to belong to the higher income deciles.On the other hand, unemployment always has a strong bias towards theless skilled — i.e. the wages of the less skilled, which can be supposed tobelong to the lower income deciles, come under the most pressure. Com-bining both trends, high income earners relative to low income earners willprofit from unemployment — which plainly means that wage dispersion andoverall income inequality will rise with growing unemployment. However,the ‘inequality elasticity’ of unemployment depends on several factors: theoverall strength of the social partners at the negotiating table (in terms ofunion density and coverage rate; see Volscho [2004]) and particularly, thecollective bargaining system, the state’s involvement in collective bargain-ing and also labour market and welfare state regulations (EU Commission2004: 109ff.). There is a clear and very strong relation between the degreeof corporatism24 of a collective bargaining system, the strictness of labourmarket regulations and the generosity of the welfare state and a solidaristicwage policy stance in terms of the compression of the qualificational wagestructure.25

These different characteristics can be sensibly clustered to form two‘models’: ‘corporatist’ and ‘liberal’ (see tab. 1). Since Esping-Andersen’s(1990) famous distinction between different ‘worlds of welfare capitalism’,many attempts have been made to group different countries. While Amable(2003) proposes as much as five different ideal types of capitalism (market-based, social democratic, Asian, continental European and South Europeanmodels), Michel Albert (1991) differentiates between an Anglo-Saxon anda Rhenish model of capitalism and Hall/Soskice (2001) identifies the liberal

24By corporatism, I not only mean the de jure centralisation of the wage bargainingsystem (i.e. at company or regional or nation-wide level) but also its de facto cooper-ation between regional and sectoral levels, union density and coverage rates and stateinvolvement. Germany, for example, is often taken at face value — i.e. wage bargain-ing struck at regional-sectoral levels, — yet the high efforts of cooperation of the tradeunions under the roof of the Deutscher Gewerkschaftsbund is not praised accordingly.Moreover France, for instance, shows low union density and de jure decentralised wagebargaining, but a very high coverage rate, a decent degree of cooperation of unions anda high state involvement in labour market regulations which would not be correctlymirrored if France would be rated at low corporatism level.

25See e.g. Barth/Zweimuller (1995), Zweimuller/Barth (1994), Moene/Wallerstein(1996) and, particularily OECD (1997) where the combinations here termed ‘corpo-ratism’ are tested. There are but two exceptions from this very close link: Austria andthe Netherlands which both manage to provide a rather wide qualificational wage dis-persion under a very centralised wage bargaining system - both countries will acquire aspecial status in this study.

366 ARNE HEISE

market economies (LME) and the coordinated market economies (CME).Without spending too much time on comparing the different categorisa-tions, I will — for the sake of simplicity and measurability and sufficientfor my limited scope — side with Albert and Hall/Soskice in broadly dis-tinguishing, at this point, a ‘corporatist bargaining model’ which is verymuch in line with Albert’s ‘Rhenish capitalism’, Hall/Soskice’s CME and a‘liberal bargaining model’ aligned with Albert’s ‘Anglo-Saxon Model’ andHall/Soskice’s LME. Renowned from several studies (see e.g. Alesina/DiTella/MacCullock 2001; Delhey 1999), we may also distinguish betweendifferent cultures of inequality which may form a second dimension to beadded to the institutional distinction: the ‘liberal culture’ allows for the ac-ceptance of high income inequality, while a ‘Scandinavian culture’ is muchmore averse to inequality and, in between, a ‘Continental culture’ of in-termediate inequality acceptance can be singled out. In tab. 1, the twodimensions are combined and those countries classified that will be empir-ically tested later.

TABLE 1.

Varieties of Capitalism in selected OECD countries

LME CME

Liberal inequality culture USA, CAN, UK, NZL, AUS

Continental inequality culture F, GER, NL, B, A, I

Scandinavian inequality culture SWE, DK, N

Source: Amable (2003); Wallerstein/Golden (2000); OECD Employment Outlook 1999;Golden/Lange/Wallerstein (2002); Delhey (1999)

The next step will be to determine a rate of unemployment, which —taking the different cultural backgrounds and institutional settings seri-ously — will best serve the interests of the meritocracy: a MeritocraticallyOptimal Rate of Unemployment (MORU). In order to do so, we assumea positive relation between unemployment and income inequality, and forthe sake of simplicity, we take the ’income dispersion function’ (ID) to belinear:

ID : Π = aU ; (1)

Π =Income dispersion as measured by D90/D10, a = institutional param-eter covering LME and CME; U = unemployment rate

The institutional parameter will be standardised as 1 in order to coverCME’s and as 2 in order to cover LME’s.

Moreover a link between the level of income inequality and the interestsof the meritocracy is established. For reasons exposed above, the high-


est income decile26 here is taken as proxi for those comprising as the elitewhich I have chosen to call ‘meritocracy’ — and the utility derived fromtheir (absolute as well as relative27) income measures the interests of themeritocracy. Furthermore there are several reasons why the utility (andeven underlying income) may eventually reach a maximum with growingincome inequality, i.e. with growing relative income: a) increasing in-equality is accompanied by growing political instability which may harmeconomic growth and (gross) income generation in general, but may alsoimpinge on crime prevention and property protection cost in particular.Thus inequality may be detrimental to the (net) income of the meritocracy(see e.g. Persson/Tabellini 1991 and Thurow 1971); b) increasing incomeinequality, simply by reducing the overall marginal propensity to consume,may reduce aggregate demand expectations and therefore reduce privateinvestment outlays and, subsequently, private consumption spending (seee.g. Heise 1999); c) research into social capital has established a clear(negative) link between income inequality and social capital in a societyand also created a clear (positive) link between the stock of social capitaland economic growth (see e.g. Halpern 2005); or d) it may, as LesterThurow (1971: 327) puts it, depend on the “aesthetic taste for equality orinequality similar in nature to a taste for paintings”.

The meritocracy’s utility function uM takes the following form:

uM = bΠ−Π2; (2)

b = cultural factor measuring inequality acceptanceHowever, the ‘optimal degree of inequality’ (i.e. where the utility func-

tion of the meritocracy reaches its maximum) will vary according to theestablished cultural embedding of individual tastes and values: again westandardise b as 3 in order to cover the Scandinavian culture, 4 to coverthe Continental culture and 5 to cover the liberal culture. The more egali-tarian values and attitudes a society professes (i.e. the lower b), the higherwill be the cost (in terms of growth detriment and crime prevention) ofgrowing income inequality and, therefore, the lower the ‘optimal degree ofinequality’. Important to notice is that this inference has been made fromthe perspective of a meritocracy embedded in a specific cultural setting.

Finally, it is only a small step to determine the ‘Meritocratically Opti-mal Rates of Unemployment’ (MORU). If there is anything like an optimalposition in personal income distribution and unemployment is a strong ex-

26It would be very reasonable to choose the highest income quartile (P75) as proxifor the meritocracy in order to widen the scope of investigation. However, if at all (seebelow) comparable data is available for only the lowest, the middle and the highestincome deciles.

27“Individuals are not just interested in their own incomes. The incomes of otherindividuals may appear in their own utility functions” (Thurow 1971: 327).

368 ARNE HEISE

planatory variable in income inequality, there must certainly be a magni-tude of unemployment which can be associated with the establishment andmaintenance of those model-specific distributions. This can be formallyproved by solving (2) under the condition of (1):

uM = b(aU)− (aU)2 (3)= baU − a2U2 (4)

The utility maximum will be determined by setting the first derivative4uM/4U = 0:

0 = ba− 2a2U (5)U = ba/2a2 (6)

In order to get the model specific Meritocratically Optimal Rates of Un-employment, we would have to insert the respective (standardised) insti-tutional and cultural parameters a and b into equation (6) as in tab. 2:

TABLE 2.

Varietes of MORUs

Bargaining model Value system Incentives

(institutional parameter a) (cultural parameter b)

High MORU

MORUconcor Corporatist Continental Low inequality

(U = 2) (centralised) elasticity + intermediate

(a = 1) (b = 4) inequality tolerance

Medium MORU

MORUscancor Corporatist Scandinavian Low inequality

(U = 3/2) -egalitarian elasticity + low


MORUliberal Liberal Liberal High inequality

(U = 5/4) elasticity + high


Low MORU

MORUcondec Corporatist Continental High inequality

(U = 1) (decentralised) elasticity + intermediate


It is the highest, if a corporatist bargaining model is coupled with a con-tinental value system showing an intermediary egalitarian stance (MORU-concor). The MORU is lowest, if a corporatist bargaining system that al-


TABLE 3.

Hypothetical equilibrium levels of unemployment and income inequalityin varieties of capitalism

MORU

Low Medium High

Scandinavian,

Low corporatist

Income CME

inequality Continental, Continental,

Culture Medium decentralised centralised

CME CME

High Liberal LME

lows for high(er) wage dispersion meets a continental value system (MORU-condec).28 Furthermore there are two intermediary MORUs if a liberalbargaining system is mixed with a liberal, non-egalitarian value system(MORUliberal) or if a corporatist bargaining system meets a Scandinavian-egalitarian value system (MORUscancor).

Before we will take a closer look at the empirical backing of the exposedpolitical economy of meritocracy, let us first formulate a hypothesis thatcan be derived from what has unfolded above (tab. 3). It can be seen thatthe Continental CMEs are sub-divided into a group of countries that arelabelled as ‘centralised’ and another group of countries — the Netherlandsand Austria — which are labelled ‘decentralised’. This distinction is madein order to take the specificity of those two countries into account which is a-typical to CME’s: a distinctly higher dispersion elasticity of unemployment(i.e. a higher parameter a). In a meritocratically optimal equilibrium, wewould expect these continental economies to have the lowest unemploymentrate which organises a high level of wage dispersion within a corporatistbargaining setting, i.e. this would be illustrated by a rather high incomeinequality. In liberal economies, we would expect a simultaneity of highincome inequality and an intermediate position with respect to unemploy-ment. Although the Scandinavian corporatist economies also would beexpected to hold an intermediate position with respect to unemployment,income inequality must be much lower here. Moreover, the corporatist-continental economies would be expected to show an intermediate positionwith respect to income inequality, yet the highest unemployment rate ofall country clusters.

28I am not describing all possible MORUs but only those which will be tested empir-ically. If there were economies that would couple a liberal bargaining system with anegalitarian value system, we would expect them to have the lowest MORU.

370 ARNE HEISE

TABLE 4.

Hypothetical trends in unemployment and inequality in varieties of capitalism

4UNR

Low Medium High

Scandinavian, Continental,

Low corporatist centralised

CME CME

Continental,

4 ID Medium decentralised

CME

High Liberal LME

For at least three reasons, we should also examine trends (i.e. changesin time) rather than exclusively levels: 1) no economy will ever be in aposition of stable equilibrium and, 2) we cannot expect the meritocracy tobe able to completely achieve its goals in terms of pushing the economy to-wards a position which would best serve their interests; 3) comparable dataon unemployment rates as well as income inequality still demonstrates ahigh degree of imponderability with respect to levels (see e.g. Moran 2005),but should be more reliable with respect to developments. Therefore, wewould have to work out what development we would expect the differentcountry clusters to take (see tab. 4) if they were to all begin from the sameposition, which would be a position of (near) full employment coupled witha rather egalitarian wage and income structure - i.e. a description which fitsthe situation very well at the end of the ’golden age’ in the 1970s. Both,in the Scandinavian-coordinated economies as well as in the Continental-centralised economies we would expect a rather low increase in incomedispersion due to the ‘sheltering’ institutions (low a). However, the latterwill have to pay a higher price — i.e. higher increase in unemployment— because the higher inequality aversion in the Scandinavian-coordinatedeconomies may render high and lasting unemployment that would be dan-gerous with respect to the interests of the Meritocracy (lower b). In theLMEs with liberal culture, unemployment would have increased quite sim-ilarly to the development in the Scandinavian-coordinated economies —as they both show intermediary MORUs — yet at much higher income in-equality dynamics. Finally, there are CMEs featuring a Continental culturethat allows for distinctly higher income dispersion (i.e. the decentralisedCME).


5. EMPIRICS OF MORU

Before we go into the empirics, let us discuss which variable we wouldideally like to look at and, under the constraints of only limited comparabledata availability, which proxy to take. First of all, it must be rememberedthat we are going to take the perspective of the meritocracy. ‘Meritocracy’has been proxied by the ‘highest income percentiles’ which is usually takenas the highest income decile (for methodological reasons, the 9th decile).However, the question arises as to which income category ought to be taken:gross or net income, individual or household income, wage or overall (in-cluding capital) income? As a meritocratic perspective is taken — i.e.earnings according to merits depending on the endowment with human, fi-nancial and real capital and its implementation within different markets —individual, overall, net income seems to be most appropriate. However, al-though differences in absolute levels of income dispersion between differentconcepts may be quite substantial and some reordering of country rankingsmay be possible (which is probably most pronounced in the case of Austria;see e.g. Pontussen 2000: 297ff. and Guger/Marterbauer 2004), the differ-ences in gross and net income varying across the different countries andcountry clusters and non-wage income seems to be more important for themeritocracy than for any other income-group in society. Furthermore, theresults are not very sensitive with respect to the income concept chosen(particularly if trends rather than levels are measured) for the followingreasons: 1) trends in disposable income distribution and individuals grossearnings are very similar (see e.g. Smeeding (2002: 195) and compare withinequality trends as displayed in tab. 5); 2) redistributing welfare systemsnarrow inequality levels but follow the exact trend of market inequality,i.e. more egalitarian economies (with respect to market (gross) income)show higher levels of redistribution (with respect to outcome in net in-comes) than less egalitarian economies29, and the trends in rising marketincome inequality are mirrored by the trends in lowered redistribution (seee.g. Kenworthy/Pontussen 2005; Burniaux 1998); 3) the most importantfactor in determining levels and trends of income inequality is by far theunderlying levels and trends of wage dispersion — and this result is robustto changes in the method of decomposing earnings into different incomecategories (see e.g. Burniaux 1998; Forster 2000).

There are basically two sources that would be able to provide interna-tionally comparable data on levels and trends of income inequality: theLuxembourg Income Study (LIS) and the OECD Labour Market Statis-tics. As the LIS provides disposable income data based on household in-come rather than individual income, this study relies on the OECD LabourMarket Statistics. Although the national definitions and sources are not

29Which has been called ‘paradox of redistribution’ by Iversen/Soskice (2004).

372 ARNE HEISE

entirely similar and the data does not cover the same time period for allcountries, this seems to be as close as we can get to standardised and com-parative data and, for the reasons outlined above, it will surely suffice withrespect not only to income trends but may also be taken as moderatelyreliable with respect to a comparison of levels.

In order to test the exposed theory of MORU, we need enough variancein unemployment and income inequality over time and among the differentcountries. As can be seen from tab. 5, at the end of the 1990s — afterthree major economic crisis — the picture of almost full employment in theearly 1970s in all countries under investigation has changed: unemploymenthas not only more than doubled to an average of 5.6%, the variance hasalso more than doubled — i.e. some countries must have coped much bet-ter than others. The same is true for income inequality (and underlyingwage dispersion): over the past two decades it has not only considerablyincreased but the texture of inequality has also become pronouncedly dif-ferent (with a much higher variance). Again, some countries have suffered(or allowed for) much more dispersion than others — and it will be ourtask to discover if there are distinguishable patterns.

In tab. 6, the results are exposed with respect to levels of unemploymentand income inequality: as expected Denmark, Sweden and Norway formingthe Scandinavian, corporatist model show only a moderate increase in un-employment and income inequality as they produce a medium MORU andare very reluctant to inequality. Canada, the United States, the UnitedKingdom, Australia and New Zealand make up the liberal model. How-ever these countries show a moderate increase in unemployment and ratherhigh change in income dispersion, while Germany, France, Italy and Bel-gium comprising the centralised continental model had to pay its only smallincrease in income inequality with a marked increase in unemployment —something which the Netherlands and Austria (taken separately to formthe decentralised continental model) were able to prevent (low unemploy-ment). This was due to an allowance for higher inequality — exactly aspredicted. The country clusters not only happen to follow exactly the hy-pothetical patterns put forward above (compare tab. 6 to tab. 3) but alsoshow only small variances — which is to say that differences among thecountries forming a cluster are low (which one would expect, if the clustermakes sense).

The picture does not change at all if we examine the trends in unem-ployment and income inequality (see tab. 7): Again, reflecting the hy-pothesis, we see a very distinct pattern of changes across the varieties ofcapitalism with, in part, extremely high degrees of coherence amongst the


TABLE 5.

Unemployment and income inequality in the early 1970s and late 1990s

UNR UNR Change Income Income Change

(mid-70) (end-90) UNR dispersion dispersion ID

(end-70) (end-90)

NL 2.9 2.5 −0.4 2.56 2.91 +0.35

DK 2.8 4.4 +1.6 2.15 2.15 +0.0

Ger 1.8 7.8 +6.0 2.88 3.04 +0.16

SWE 2.0 4.9 +2.9 2.03 2.22 +0.19

NOR 1.5 3.3 +1.8 2.05 2.08 +0.03

F 2.8 8.5 +5.7 3.25 3.05 −0.20

B 2.3 6.7 +4.4 2.30 2.20 −0.10

I 5.0 10.4 +5.4 2.32 2.40 +0.08

A 1.3 3.6 +2.3 3.45 3.66 +0.21

USA 5.6 4.0 −1.6 3.78 4.57 +0.79

CAN 5.0 6.8 +1.8 4.02 4.17 +0.15

UK 2.0 5.0 +3.0 2.91 3.45 +0.54

NZL 0.5 3.9 +3.4 2.89 3.41 +0.52

AUS 2.7 6.3 +3.6 2.73 2.87 +0.14

(unweighted) 2.7 5.6 +2.8 2.81 3.01 +0.20

Average

Variance 2.2 5.08 +4.79 0.399 0.603 +0.204

Note: ‘mid-70’ refers to the respective year with lowest unemployment in the first half of the1970s; ‘end-70’ refers to the last year before inequality started to rise — never later than1979; ‘end-90’ refers to the end of the business cycle in the late 1990s/early first decade of the21st century (unemployment rates) and respective year for income dispersion (if available;otherwise the latest available year is taken).Source: UNR = internationally standardised unemployment rates: OECD (2005); Incomedispersion = 9th decile/ 1st decile; daily, weekly or monthly gross earnings: OECD-LabourMarket Statistics DATA and OECD (1996) for Austria and Norway. For Norway, data isonly available for 1991. See appendix: earnings documentation.

374 ARNE HEISE

TABLE 6.

Unemployment and income inequality at the end of the 1990s in varietiesof capitalism; absolute levels

MORU

Low Medium High

(SWE; DK; Nor)

Low UNR:4.2 [0.67]

ID: 2.15 [0.01]

Income (A; NL) (Ger; F; B; I)

inequality Medium UNR: 3.0 [0.605] UNR: 8.4 [2.4]

ID: 3.28 [0.281] ID: 2.67 [0.192]

(USA; UK; AUS;

NZL, CAN)

High UNR: 5.2 [1.7]

ID: 3.69 [0.453]

Notes: Italic numbers in square brackets show the variance among country clusters. Theoverall variance in unemployment rates is 5.08 and in income dispersion 0.603.Source: See tab. 5

country clusters30 — i.e. the overall variance is an inter-model variance,not an intra-model variance. The results are clear: Scandinavian, corpo-ratist economies are more successful in terms of unemployment increasesas they are, on the one hand, less tolerant to inequality and, on the otherhand, have allowed for a slightly stronger increase in inequality than theContinental, centralised countries. The liberal economies experienced astrong rise in inequality which was triggered by a modest increase in un-employment, while the Continental decentralised economies paid for thegood labour market performance by an intermediary rise in income dis-persion despite its low tolerance for inequality (as compared to the liberalmodel).

To make the empirical picture complete, let us come back to Okun’s ‘bigtrade off’ mentioned at the beginning. In tab. 8, equations (2) and (3) showthat there is definitely no statistically significant correlation between thelevel of unemployment and the level of income inequality but only a rather

30The rather poor performance of the continental, decentralised cluster with respectto variance in unemployment trends is not really a surprise as we only have a sampleof two countries. Furthermore, in the case of the liberal economies, the rather highvariance stems only from the exceptionally well development of unemployment in theUS as compared to the early 1970s which is due to a number of very distinct factors. Onlyone of these factors is that unemployment rates in the USA are seriously underestimatedas a large part of long-term unemployment is represented through prison inmate figuresrather than unemployment figures; see e.g. Katz/Krueger (1999) or Western/Beckett(1999). If the data was corrected for these underestimations, the variance would dropto below 1.0.


TABLE 7.

Changes in unemployment and income inequality in varieties of capitalism(absolute change since 70s)

4 UNR

Low Medium High

(SWE; DK; Nor) (Ger; F; B; I)

Low UNR: +2.1 [0.490] UNR: +5.4 [0.482]

ID: +0.04 [0.018] ID: −0.01 [0.028]

(A; NL)

4 ID Medium UNR: +1.0 [3.645]

ID: +0.28 [0.008]

(USA; UK; AUS;

NZL, CAN)

High UNR: +2.0 [4.628]

ID: +0.42 [0.086]

Notes: Ialic numbers in square brackets show the variance among country clusters. Theoverall variance in changes of unemployment rates is 4.79 and in changes in income dis-persion 0.204.Source: See tab. 5

weak, yet significant correlation between the change in income inequalityand the change in unemployment (regressions 5 and 6) — which is usuallytaken as argument for the quest for higher income differentiation as a pre-requisite for higher employment growth.31 If the inverse causation runningfrom unemployment (changes and levels) to inequality (changes and lev-els) suggested by the exposed MORU theory was correct, the rather weakand insignificant correlations would depend on the model-specific varia-tions resulting in non-linear relations between dependent and independentvariables. Once this is taken into account by introducing a simple variablecatching the affiliation of individual countries to specific country clusters,we would expect a clear rise in the strength and significance of the rela-tionships. As we can see from regressions 1 and 4 in tab. 8, this is actuallythe case — rendering not only the independent variables ‘change in unem-ployment rates’ significant (in equation 1 and more significant in equation4), but also the variable ‘Model’ and the whole specification (see F test)and it reduces clearly the risk of misspecified functions (as in equations 5and 6).

31As the International Monetary Fund (IMF) and the Organisation for EconomicCooperation and Development (OECD) play leading roles in propagating the ‘big tradeoff’, this is sometimes called ‘IMF-OECD orthodoxy’; see e.g. Siebert (1997); IMF(2003); OECD (1997).

376 ARNE HEISE

TABLE 8.

OLS regression results — Inequality and Unemployment

(1) (2) (3) (4) (5) (6)

IDEnd90 IDEnd90 IDEnd90 4 ID 4 ID 4 ID

UNREnd90 −.041 −.055∗4 UNR −.173∗ −.138 −.079 ∗ ∗∗ −.069 ∗ ∗Model .376 ∗ ∗ .106∗R-Square .425 .151 .018 .517 .329 .220

F test 4.063 ∗ ∗ 2.142 .219 5.876 ∗ ∗ 5.877 ∗ ∗ 3.427∗DW test 2.282 1.515 1.390 1.869 1.033 .937

N 14 14 14 14 14 14

Notes: IDEnd90 = P90/P10−ratio at the end of the 1990s; 4ID = absolute change inincome distribution between the end of the 1970s and the end of the 1990s; UNREnd90= internationally standardized unemployment rate at the end of the 1990s; 4 UNR =absolute change in unemployment rates between early 1970s and late 1990s; Model = variablecatching different model affiliation; ∗ ∗ ∗ =significant at 1% error level; ∗∗ =significant atthe 5% error level, ∗ =significant at the 10% error level; DW test = Durbin−Watson test onautocorrelation; N = number of cases.Source: See tab. 5.

6. CONCLUSION: MORU AND THE POLICY OF FULLEMPLOYMENT IN THE EU

The empirics are so strong that the proposed political economy of mer-itocracy cannot easily be ignored. However, this is not the place for aconspiracy theory of unemployment. It would certainly be oversimplifyingto argue that the political actor — i.e. governments and central banks —care only for meritocratic interests. Even if the very individuals who areacting on behalf of the political actor were taken as part of the meritoc-racy, in a democratic electoral environment it would be difficult for themto outspokenly pursue their interests if they cared about being re-elected.Therefore, meritocratic interest can only be pursued if the interests havebeen successfully put forward and connotated as ‘common good’. Withinthe confines of this article, it is impossible to show how this common goodcan be achieved in general, and has been achieved in the particular coun-tries under investigation here. What seems obvious is that the meritocracyas a group is too amorphous to pursue its vested interests by way of or-ganising a pressure group.32 Therefore, it seems much more promising tofollow Gramscian lines of hegemony in order to understand how ideolo-

32Which does not rule out the possibility of ideological networks to be formed thatmay act as pressure group: The Mont Pelerin Society is an organisation which worksin secrecy but has a strong world wide influence on agenda building and agenda settingprocesses; see e.g. Desai 1994, Collard 1968).


gies favouring certain interests may and have become dominant (see e.g.Cockett 1995; Cox 1996; van der Pijl 1995).

The instrumental transmission mechanism of meritocratic interests ismore obvious: again, Michal Kalecki foresaw already as early as the mid-1940s how meritocratic politics will nevertheless prevail, even if the meri-tocracy represents only a tiny minority of society and their interests standagainst the interests of the general public welfare: By way of de-legitimizingor de-capabilising Keynesian macropolicies33 in academic and public dis-courses and enthroning orthodox policies of ‘sound money’ and ‘sound fi-nance’34, the economic policy maker gets the instruments of coarse-tuningmeritocratic unemployment at hand whether he is aware of it or not. Byway of ‘under-cover, hidden’ Keynesian macropolicies Kalecki neverthelesshas instruments to fine-tune unemployment if the respective MORU is closeat reach35. Furthermore, academics fail to address the obvious discrepan-cies between outspokenly neo-liberal, supply-side advocates (‘read my lips’)and the very same having pursued a macropolicy which most closely re-sembles Keynesianism: the Reagan and George W. Bush administrationsin the USA and the Thatcher administration in the UK. What seems tobe a puzzle and can, therefore, only be addressed as ‘ironic’, as has beendone by Norbert Andel in his presidential speech to the International In-

33In a first round, the ‘monetarist counter-revolution’ made Keynesian policies ac-countable for the stagflation problems and growing public indebtness of the 1980s and1990s. In a second round, globalisation has been claimed to undermine the capabilitiesof Keynesian policy making at national levels. Both blows together were severe enoughto turn even progressive, non-economic social scientists and social democratic policymakers away from Keynesian demand management; see e.g. Scharpf (1991); Iversen(1999).

34The actual transmission from interests to ideology and then to public policy inmodern media democracies, have been presented by this author against the backgroundof German Social Democratic Politics; see Heise (2005b).

35Which is a very important part of the story as the political situation may easilybecome unstable if growing inequality (and poverty) would be paralleled with risingunemployment. This can be seen for example through the infamous ‘Toxteth riots’ atthe beginning of the Thatcher era when unemployment soared and inequality began torise dramatically. The meritocracy has no interests in surpassing ‘its MORU’. Again,the actual making of the policies must not necessarily be downright meritocratic, butmay, for instance, come about by a ‘pragmatic’ monetary policy taking unemployment(or the ‘output gap’) in a symmetric way into consideration (i.e. the policy reactionfunction includes output gaps just as much as inflation gaps). This may explain why themonetary policy of the FED is typically regarded as more ‘pragmatic’ than the monetarypolicy of the ECD or its forerunner, the Deutsche Bundesbank (see e.g. Fritsche et al.2005 or Siklos 2004). On the other hand, the anti-inflation bias associated with ‘soundmonetary policy’ plays a crucial role in keeping up pressure on inequality (directly andvia unemployment; see e.g. Lippi 1999: 63ff.).

378 ARNE HEISE

stitute of Public Economics36, is perfectly understandable in the contextof a political economy of meritocracy.

What are the consequences of the political economy of meritocracy ex-posed above? Firstly, it is important to address the interests and hiddenagenda involved. Quite as it was Mancur Olson’s approach, in order to letthe public welfare prevail, distributional interests must be rejected and ‘dis-tributional coalitions’ abolished. This, however, is not as easy as it soundsbecause the meritocracy does not ‘seek its rents’ through organisations thatact as monopolies on markets, but rather through elite pressure that mo-nopolises public opinion (‘pensee unique’). In such a state, only educatingthe public can work, and a less uniform ideology of mass political parties(i.e. veritable ‘political competition’) would help. Secondly, one may betempted to think of reducing the inequality tolerance of society by, for in-stance, a collusion of those suffering most from the policy of meritocracy:the low skilled unemployed. However, this may result in a politically verycostly and probably uncontrollable path if it helps radical parties to settleand grow in the political spectrum. Thirdly, one might think of joining thecontinental, decentralised model by increasing the inequality elasticity ofunemployment. However, there are also drawbacks to this solution: On theone hand, there seems to be a trade off between inequality and efficiencyin terms of productivity growth. An egalitarian wage policy has, for in-stance, always served as ‘productivity whip’ in Sweden and a productivityslowdown is alleged to have accompanied the good employment perfor-mance of the Netherlands (see Naastepad/Kleinknecht 2004; Kleinknecht2003). On the other hand, dispersing the wage scale, particularly at thelow skilled end, may reduce employer’s incentives to train the workforceand can severely harm production systems that are based on vocationaleducation (e.g. Germany; see Rogers Hollingsworth 2000: 284ff.). Howevermost importantly, this solution is based on the capability of the politicalactor to use monetary and fiscal policy measures in order to fine tune theeconomy (although this will not necessarily be the announced policy). If,albeit, governments have ‘tied their hands’ to principles of ‘sound money’and ‘sound finance’ not only ideologically but also institutionally — as isthe case in the European Union with the independence of the EuropeanCentral Bank and the establishment of the ‘Stability and Growth Pact’ —,they may find it difficult to pursue the kind of expansionary macropolicyneeded to reward growing wage dispersion and income inequality.

We have to face reality: in economic particulars, everything is directedtowards the distribution of outcomes. This is the case in direct economicencounters on markets. It is also the case in societal encounters most di-

36“It is ironic that in the same country where the counterrevolution against the key-nesian revolution originated, the conservative government uses dept finance to an extentwithout any parallel in the time of peace” (Andel 1986: 11).


rectly through social policies and, of course, through economic policies ingeneral. To believe, as Keynes seemed to, that there are ‘objective’ eco-nomic policies based on the ‘correct’ understanding of the ‘real economicworld’ and which are ‘functionally’ addressed to cure economic problems(as the dentist cures tooth aches37) does not take into account that eco-nomic policy is embedded within a political system that must accommodatediverse interests. Mancur Olson (1996: 92) concluded a summing up of histheory in the following:

“The most important implication of the analysis, however, is that the onlyreal solution is for societies to acquire a better understanding of economicsand of the present argument. ... No historical process that is understood isinevitable.”

Despite my disagreement with Olson’s particular approach and my alter-native way of looking at politico-economic matters, this general statementcan certainly and wholeheartedly be embraced.

REFERENCESAlbert, M., 1991. Capitalisme Contre Capitalisme. Paris: Edition du Seuil.

Alesina, A., R. Di Tella, and R. MacCullock, 2001. Inequality and happiness: AreEuropean and American different? CEPR Discussion Paper 2877.

Amable, B., 2003. The Diversity of Modern Capitalism. Oxford.

Andel, N., 1986. Changing concepts of public debt in the history of economic thought.In: Public Finance and Public Debt. Proceedings of the 40th Congress of the IIPF.Ed. by B. P. Herber. Detroit: Wayne State University Press, 1-13.

Arestis, P. and M. Sawyer, 1998. Keynesian economic policies for the new millennium.Economic Journal 108, 181-195.

Baldwin, P., 1990. The Politics of Social Solidarity: Class Bases of the EuropeanWelfare State. 1875 - 1975, Cambridge: Cambridge University Press.

Barth, E. and J. Zweimuller, 1995. Relative wages under decentralized and corporatistbargaining systems. Scandinavian Journal of Economics 97, 369-384.

Benabou, R., 2000. Meritocracy, redistribution, and the size of the pie. In: K. J.Arrows (ed.), Meritocracy and Economic Inequality. Princeton: Princeton UniversityPress, 317-339.

Boddy, R. and J. Crotty, 1974. Class conflict and macropolicy: The political businesscycle. Review of Radical Political Economy 7, 1-19.

Bowles, S., R. Edwards, and F. Roosevelt, 2005. Understanding Capitalism: Compe-tition, Command, and Change. Oxford: Oxford University Press.

Burklin, W. et al. (eds.), 1997. Eliten in Deutschland. Rekrutierung und Integration,Opladen: Westdeutscher Verlag.

37“But, chiefly, do not let us overestimate the importance of the economic problem,or sacrifice to its supposed necessities other matters of greater and more permanentsignificance. It should be a matter for specialists — like dentistry. If economists couldmanage to get themselves thought of as humble, competent people, on a level withdentists, that would be splendid” (Keynes 1930/1972: 332).

380 ARNE HEISE

Burniaux, J.-M. et al., 1998. Income Distribution and Poverty in Selected OECDcountries. OECD Economic Department Working Papers 189.

Cockett, R., 1995. Thinking the Unthinkable. Think-Tanks and the EconomicCounter-Revolution 1931 - 1983. London: Fontana Press.

Collard, D., 1968. The New Right: A Critique. London: Fabian Society.

Cornwall, J. and W. Cornwall, 2005. Power and institutions in macroeconomic theory.Cambridge Centre of Economic and Public Policy Working Paper 02/05.

Cox, R.W., 1996. Ideologies and the new international economic order: reflections onsome recent literature. In: Cox, R.W. and T.J. Sinclair (eds.). Approaches to WorldOrder. Cambridge: Cambridge University Press, 376-419.

Davis, St. and Chr. Meyer, 2000. Future Wealth. Cambridge: MIT Press.

Delhey, J., 1999. Inequality and attitudes, postcommunism, western capitalism, andbeyond. WZB Discussion Paper FS III/99-403.

Desai, R. 1994. Second-hand dealers in ideas: Think-tanks and the thacherite hege-mony. New Left Review 203, 27 - 64.

Esping-Andersen, G., 1990. The Three Worlds of Welfare Capitalism. Princeton:Princeton University Press.

EU Commission, 2004. Employment in Europe 2003, Luxembourg : Office for OfficialPublications of the European Communities.

Forster, M.F., 2000. Trends and driving factors in income distribution and poverty inthe OECD Area. OECD Social Policies Studies Division, Labour Market and SocialPolicy Occasional Paper 42.

Fontana Dictionary of Modern Thought. 1977. Ed. by Alan Bullock. London: FontanaPress.

Fritsche, U. et al., 2005. Wirtschaftspolitische Regime westlicher Industrienationen.Unterschiede, Wachstumsperspektiven und wirtschaftspolitische Optionen in aus-gewahlten Landern, Abschlussbericht eines Forschungsprojektes der Hans-Bockler-Stiftung, Berlin unpublished manuscript).

Galbraith, J.K., 1997. Dangerous metaphor: The fiction of the labor market. TheJerome Levy Institute of Bard College Public Policy Brief 36.

Galbraith, J.K., 1998. Created Unequal: The Crisis in American Pay. New York: FreePress.

Giersch, H., K.-H. Paque, and H. Schmieding, 1992. The Fading Miracle: FourDecades of Market Economy in Germany. Cambridge: Cambridge University Press.

Gilder, G., 1982. Wealth and Poverty. London: Basic Books.

Golden, M., Lange, P. and M. Wallerstein, 2002. Union centralization among advancedindustrial societies: an empirical study.http://www.shelly.polisci.ucla.edu/data.

Guger, A., and M. Marterbauer, 2004. Die langfristige entwicklung der einkom-mensverteilung in osterreich. WiFo-Studie im Auftrag des Bundesministeriums fursoziale Sicherheit, Generationen und Konsumentenschutz, Wien: WiFo.

Hall, P. and D. Soskice (eds.), 2001. Varieties of Capitalism: The Institutional Foun-dations of Comparative Advantage. Oxford: Oxford University Press.

Halpern, D., 2005. A matter of respect. Prospect 112, 40-43.

Harrison, B., and B. Bluestone, 1990. The Great U-Turn: Corporate Restructuringand the Polarizing of America New York: Basic Books.


Heise, A., 1999. Grenzen der Deregulierung. Institutioneller und struktureller Wandelin Großbritannien und Deutschland, Berlin: Sigma.

Heise, A., 2001. New Politics. Integrative Wirtschaftspolitik fur das 21. Jahrhundert,Munster: Westfalisches Dampfboot.

Heise, A., 2005a. Limitations to Keynesian demand management through monetarypolicy — Whether Cartesian policy control, Hamburg University of Economics andPolitical Science Working Papers on Economic Government 14.

Heise, A., 2005b. German social democratic economic politics in the light of agendatheory. Intervention. Journal of Economics 2, 131 - 151.

Henley, A., 1988. Political aspects of full employment: A reassessment of kalecki.Political Quarterly 59, 437 - 450.

Henry, W., 1995. In Defence of Elitism. New York: First Anchor Books.

IMF, 1999. Chronic unemployment in the euro area: causes and cures. In: WorldEconomic Outlook. Washington: IMF.

Iversen, T., 1999. Contested Economic Institutions. The Politics of Macroeconomicsand Wage Bargaining in Advanced Democracies. Cambridge: Cambridge UniversityPress.

Iversen, T. and D. Soskice, 2004. Electoral system and the politics of coalition. Paperpresented to the Conference of Europeanists. Chicago.

Jantti, M. and S. R. Jenkins, 2001. Examining the impact of macro-economic condi-tions on income inequality. Working Paper of the Institute of Social and EconomicResearch, University of Essex 17.

Kalecki, M., 1943/1990. Political aspects of full employment. Political Quarterly 14,322 - 331; cited after: Kalecki, M.; Collected Works of Michal Kalcki: Vol. I: Capital-ism. Business Cycles and Full Employment, ed. by Jerzy Osiatynski, Oxford: OxfordUniversity Press, 347 - 356.

Katz, L. and A. Krueger, 1999. The high-pressure US labour market of the 1990s.Brookings Papers on Economic Activity 1, 1-87.

Kekes, J., 2003. The Illusions of Egalitarianism. Ithaca: Cornell University Press.

Kenworthy, L. and J. Pontussen, 2005. Rising inequality and the politics of redistri-bution in affluent countries. LIS Working Paper 400.

Keynes, J.M., 1930. Economic possibilities for our grandchildren, nation andathenaeum. Reprinted in: Keynes, J.M.; The Collected Writings of John MaynardKeynes. Vol. 9: Essays in Persuasion, London 1972, 321-332.

Kleinknecht, A., 2003. Causes of the Dutch job miracle: There is no free lunch. DeEconomist 151, 329-333.

Kotz, D.M., T. McDonough, and M., Reich (eds.), 1994. Social Structures of Ac-cumulation: The Political Economy of Growth and Crisis. Cambridge: CambridgeUniversity Press.

Lembruch, G., 1984. Concertation and the structure of corporatist networks. In J.Goldthorpe (ed.). Order and Conflict in Contemporary Capitalism. Oxford: OxfordUniversity Press, 60-80.

Lippi, F., 1999. Central Bank Independence, Targets and Credibility. Cheltenham:Edward Elgar.

Machan, T.R. (ed.), 2002. Liberty and Equality. Washington: Hoover InstitutionPress.

382 ARNE HEISE

Marglin, S., Schor, J.(eds.), 1990. The Golden Age of Capitalism. Oxford: OxfordUniversity Press.

Marin, B., 1983. Organising interests by interest associations: Organizational pre-requisites of co-operation in Austria. International Political Science Review 4, 197-216.

Moene, K.O. and M. Wallerstein, 1996. Egalitarian wage policies and economic per-formance. Paper presented at the annual meeting of the American Political ScienceAssociation, San Francisco, August 28 - September 1.

Moran, T.P., 2005. Bootstrapping the LIS: Statistical inference and patterns of in-equality in the global north. LIS Working Paper 378.

Naastepad, C.W.M. and A. Kleinknecht, 2004. The Dutch productivity slowdown:The culprit at last? Structural Change and Economic Dynamics 15, 137-163.

Nordhaus, W.D., 1975. The political business cycle. Review of Economic Studies 42,169-190.

OECD, 1996. Employment Outlook 1996. Paris: OECD.



Okun, A.M., 1970. The Political Economy of Prosperity. Washington: Brookings In-stitution.

Okun, A.M., 1975. Equality and Efficiency: The Big Tradeoff. Washington: BrookingsInstitution.

Olson, M., 1965. The Logic of Collective Action. Cambridge (Mass.): MIT Press.

Olson, M., 1982. The Rise and Decline of Nations. New Haven: Yale University Press.

Olson, M., 1996. The varieties of Eurosclerosis: The rise and decline of nations since1982. In: Crafts, N. and G. Toniolo (eds.). Economic Growth in Europe since 1945.Cambridge: Cambridge University Press, 73-94.

Persson, T. and G. Tabellini, 1991. Is inequality harmful for growth: theory andevidence. CEPR Discussion Paper 581.

Pontussen, J., 2000. Labor market institutions and wage distribution. In: Iversen,T., Pontussen, J. and D. Soskice (eds.). Unions, Employers, and Central Banks.Macroeconomic Coordination and Institutional Change in Social Market Economies.Cambridge: Cambridge University Press, 292-330.

Riese, H., 1981. Theorie der produktion und einkommensverteilung. Kyklos 34, 540-562.

Robinson, J., 1976. Michal Kalecki: A neglected prophet. New York Review of Books.4th of March.

Rogers Hollingsworth, J., 2000. Continuities and changes in social systems of produc-tion: The cases of Japan, Germany, and the United States. In: Rogers Hollingsworth,J. and R. Boyer (eds.). Contemporary Capitalism. The Embeddedness of Institutions.Cambridge: Cambridge University Press, 265-310.

Scharpf, F.W., 1991. Crisis and Choice in European Social Democracy. Ithaca: Cor-nell University Press.

Schettkat, R. and E. Appelbaum, 1996. The importance of labor market institutionsfor economic development. In: Gerlach, K. and R. Schettkat (eds.). Beitrage zurneukeynesianischen Makrookonomik. Berlin: Sigma, 135-155.


Siebert, H., 1997. Labor market rigidities: At the root of unemployment in Europe.Journal of Economic Perspectives 11, 37-54.

Siklos, P. L., 2004. Central bank behaviour, the institutional framework, and policyregimes: Inflation versus noninflation targeting countries. Contemporary EconomicPolicy 22, 331-343.

Smeeding, T.M., 2002. Globalisation, inequality and the rich countries of the G-20:Evidence from the luxembourg income study (LIS). In: Gruen, D., O’Brien, T. andJ. Lawson (eds.). Globalisation, Living Standards and Inequality. Proceedings fromthe annual conference of the Reserve Bank of Australia Economic Department heldin Sydney on 27 - 29 May 2002, 179-206.

Soskice, D., 1999. Divergent production regimes: Coordinated and uncoordinatedmarket economies in the 1980s and 1990s. In: Kitschelt, H. et al. (eds.). Continuityand Change in Contemporary Capitalism. Cambridge: Cambridge University Press,101-134.

Sraffa, P., 1960. Production of Commodities by Means of Commodities. Cambridge:Cambridge University Press.

Thurow, L., 1971. The income distribution as a pure public good. Quarterly Journalof Economics 83, 327-336.

van der Pijl, K., 1995. The second glorious revolution: globalizing elites and historicalchange. In: Cox, R.W. and B. Hettne (ed.). International Political Economy. NovaScotia: Fernwood Publishers, 100-128.

Volscho, T.W., 2004. Income distribution in 14 OECD nations, 1967 - 2000: Evidencefrom the Luxembourg income study. LIS Working Paper 386.

Wallerstein, I. and M. Golden, 2000. Postwar wage setting in the Nordic countries.In: Iversen, T., Pontusson, J. and D. Soskice (eds.). Unions, Employers, and Cen-tral Banks. Macroeconomic Coordination and Institutional Change in Social MarketEconomies. Cambridge: Cambridge University Press, 107-137.

Western, B. and K. Beckett, 1999. How unregulated is the US labor market? Thepenal system as a labor market institution. American Journal of Sociology 104, 46-76.

Young, M., 1958. The Rise of Meritocracy 1870-2033. London: Penguin Books.

Zweimuller, J. and E. Barth, 1994. Bargaining structure, wage determination, andwage dispersion in 6 OECD countries. Kyklos 47, 81-93.

a post-kaleckian, post-olsonian approach to unemployment and...

Documents